Some software stocks may be even less desirable in a rebounding market

Some recent developments suggest the beleaguered software group of stocks may be coming back into favor, as I noted a couple weeks ago following positive responses to both Snowflake and Elastic.

The interest in software is showing up again this week, with initiations and upgrades and downgrades on a raft of names.

NOT EVERYONE’S A WINNER

First, the bad news: Some software stocks may be even less desirable in a rebounding software market.

Wednesday, J.P. Morgan’s Pinjalim Bora rejiggered ratings on three software stocks: C3.ai, which just reported positively on Monday; PagerDuty, which sold off last month on its rather mixed earnings report; and Sprinklr, which also had a downbeat report last week.

In contrast to the exuberance around C3 at most shops, Bora is not impressed. He cut his rating to Underweight from Neutral, as the company’s “uneven and subpar growth-plus-margin performance leaves a lot to be desired, and it remains a big outlier compared to a broader base of peers trading at over 10 times FTM revenue.”

I think Bora pretty much summed up what I said on Monday in being underwhelmed: tepid growth, rich stock price.

PagerDuty is a little more surprising. Bora cuts his rating to Underweight as well because of rising competition. In case you don’t remember, PagerDuty makes software that allows a company’s IT team to know when computer systems are having issues, such as a system failure, and the software helps them take steps to remediate it.

Lots of companies are coming into the market, threatening PagerDuty’s customer acquisition and potentially putting pressure on its pricing control, writes Bora.

“We are observing a commoditization of the core on-call scheduling capabilities in the market,” he writes. “We are seeing large established vendors come into this space recently, in addition to the emergence of a large swath of startups in the space, most of which are entering the market as cheaper alternatives to PagerDuty.”

Many of these new entrants are banking on the idea that “AIOps,” a focus on artificial intelligence and automation, is taking over PagerDuty’s market, Bora notes.

For example, Bora says Datadog rolled out something competitive this summer called “On-Call,” and in the most recent earnings call the company said it was getting a “very strong reception . . . as customers are requesting On-Call as part of their deal.”

Likewise, software maker Atlassian is directing customers to use its JIRA software for IT. There’s also progress being made in the open-source world, including something called Grafana. Bora notes an “explosion” of startups, as “the number of Incident Response vendors has gone from ~70 in the beginning of 2022 to ~100 as of now, with the number of vendors serving the enterprise doubling in that time period from 15 to 30.”

For Sprinklr, Bora cuts his rating to Neutral from Overweight. The stock’s valuation is “undemanding at current levels,” he writes, namely, two times next year’s likely revenue. Still, he’s “moving to the sidelines” because the growth just isn’t there at the moment. “We await the emergence of a sustainable growth plus margin glide path over the next 12 to 18 months,” he says.

SOME LONG-AWAITED AI PAYOFF

Thursday, Macquarie Research’s Steve Koenig started coverage of six companies, including Datadog and Atlassian, but also Salesforce, Autodesk, GitLab, and MongoDB.

Koenig’s overall theme is that these vendors are poised to finally get some of the artificial intelligence payoff investors have been waiting for.

He sees the “accelerating cloud revenue trends at major hyperscalers” as something that “bodes well for software, as customers have largely worked their way through cloud spending optimization efforts and are investing in digital transformation, cloud migration, and AI.”

AI is no longer just about Nvidia’s chips: “AI investments are flowing upwards in the stack, from GPUs to infrastructure-as-a-service (IaaS) to platform-as-a-service (PaaS) and applications,” writes Koenig, referring to two cloud-based software categories.

Koenig likes Autodesk, Datadog, and GitLab, rating them each Outperform, and slaps a Neutral rating on the other three.

GitLab is Koenig’s “top pick” based on “our perception of a significant valuation discount and catalysts that could include FY26 revenue outperformance, margin expansion, improving Rule of 40 scores, and strong FCF generation.” The company is past the worst of the macroeconomic headwinds in its market for developer tools, he writes. And the valuation, at 10.8 times the next 12 months’ projected revenue, is a 28% discount to what he thinks it should be,  which is 13.8 times. That is “the seventh-largest discount in our 40-company growth software index,” he notes.

Datadog has “near-term catalysts that could include significant FY25E revenue outperformance as the year progresses,” Koenig says. For instance, the company’s “moves to embrace AI provide exposure to large language model (LLM) inferencing (with LLM observability) and differentiate its workflow and analysis capabilities.” He sees the positives at Datadog driving “top- and bottom-line outperformance, with revenue growth above the 20% range through FY27E accompanied by incremental margin expansion.”

Some of the positive stuff is merely accounting changes. Autodesk, Koenig writes, “has not lived up to its profit potential,” and its free cash flow has been “volatile” for several years. However, he is optimistic the fundamentals will improve with a new transaction model that Autodesk has undertaken.

In that new model, the company is recording as revenue what it used to pay resellers, by selling directly to the end customer. That could boost the company’s “billings” next year, the non-GAAP number for what it brings in each quarter, though the exact amount, according to Koenig, is “opaque.” More important, he says, selling direct will “help ADSK optimize its operations over time, potentially enhancing profitability,” via a more direct path to the customer.

SHORT TAKES: PALANTIR’S EXPENSIVE, REDDIT’S GOT UPSIDE

There were a few other scattered notes of coverage here and there.

Shadowy spook software maker Palantir has a lot to recommend it, but not its valuation, writes William Power of RW Baird on Thursday, initiating coverage of the stock with a Neutral rating. On the one hand, the company is “catalyzing growth” with AI via its artificial intelligence platform introduced last year, Power observes, also noting that the company “has excelled at actually putting generative AI applications into production, which is where we expect most value to be extracted in the coming years.”

But after more than quadrupling in value this year, the shares trade for 48 times revenue, way, way above most cloud software stocks, which trade for 16 times on average. Even if the company keeps growing revenue by 30% through 2026, its multiple would still be 35 times.

Power gives the stock a $70 price target, just below Thursday’s $73.20, which would equate to 40 times 2026 revenue.

I think Power is being too generous. The valuation for Palantir is way too much, driven by this year’s trading momentum. We saw this with Snowflake in 2020, when it went public and had the highest software multiple ever. Snowflake shares are down 8% since that IPO—four years later. It’ll be the same for Palantir, I’ll bet: It won’t grow into this elevated valuation.

Another one this week was Wells Fargo’s Ken Gawrelski’s initiation of social-media property Reddit with an Overweight rating and a $206 price target—26% upside from a recent $163.51.

Its stock is also expensive, but justifiably so, he believes. “We acknowledge the premium multiple (~30X ’26 EBITDA), but believe it comes with premium growth (30% 3yr Rev CAGR).”

There is “ample near-term opportunity for RDDT to enhance ad monetization,” Gawrelski says, with the company making only a third of what Meta Platforms makes in terms of advertising dollars per minute of usage, which is currently $5.60 for every 1,000 minutes of usage, by estimate, versus $17.98 for Meta. Reddit can boost that, he argues, by increasing the number of ads shown and introducing new ad formats.

“Ad load optimization” (boosting the number of ads) is “an obvious opportunity for Reddit to take advantage of its fast-growing engagement base” to increase monetization, Gawrelski writes. Data he’s seen suggest the company “currently has the lowest ad load compared to Meta, Pinterest, and Snap, at “1.5 ad impressions per minute” versus two for Meta, and three to four for the others.

If Reddit can boost that monetization in the next couple of years by just 5 percentage points, to 35% of Meta’s monetization, Gawrelski believes it could mean another $238 million in revenue in 2026 above what’s already expected by the Street.

In my view, it’s a modest proposition on Gawrelski’s part, but it’s also a slim prospect on which to hang a very expensive stock multiple.

I don’t have space here to go into the others, but Nutanix was started at a Buy rating at UBS on Thursday, and Box was started with a Buy rating at Merrill Lynch.

To me, these are all signs that whichever the stock, the software appetite is rising.

This story originally appeared in The Technology Letter and is republished here with permission.